Research progress in water quality prediction based on deep learning technology: a review.

Water, an invaluable and non-renewable resource, plays an indispensable role in human survival and societal development. Accurate forecasting of water quality involves early identification of future pollutant concentrations and water quality indices, enabling evidence-based decision-making and targeted environmental interventions. The emergence of advanced computational technologies, particularly deep learning, has garnered considerable interest among researchers for applications in water quality prediction because of its robust data analytics capabilities. This article comprehensively reviews the deployment of deep learning methodologies in water quality forecasting, encompassing single-model and mixed-model approaches. Additionally, we delineate optimization strategies, data fusion techniques, and other factors influencing the efficacy of deep learning-based water quality prediction models, because understanding and mastering these factors are crucial for accurate water quality prediction. Although challenges such as data scarcity, long-term prediction accuracy, and limited deployments of large-scale models persist, future research aims to address these limitations by refining prediction algorithms, leveraging high-dimensional datasets, evaluating model performance, and broadening large-scale model application. These efforts contribute to precise water resource management and environmental conservation.

Super-resolution reconstruction, recognition, and evaluation of laser confocal images of hyperaccumulator Solanum nigrum endocytosis vesicles based on deep learning: Comparative study of SRGAN and SRResNet.

It is difficult for laser scanning confocal microscopy to obtain high- or ultra-high-resolution laser confocal images directly, which affects the deep mining and use of the embedded information in laser confocal images and forms a technical bottleneck in the in-depth exploration of the microscopic physiological and biochemical processes of plants. The super-resolution reconstruction model (SRGAN), which is based on a generative adversarial network and super-resolution reconstruction model (SRResNet), which is based on a residual network, was used to obtain single and secondary super-resolution reconstruction images of laser confocal images of the root cells of the hyperaccumulator Solanum nigrum. Using the peak signal-to-noise ratio (PSNR), structural similarity (SSIM) and mean opinion score (MOS), the models were evaluated by the image effects after reconstruction and were applied to the recognition of endocytic vesicles in Solanum nigrum root cells. The results showed that the single reconstruction and the secondary reconstruction of SRGAN and SRResNet improved the resolution of laser confocal images. PSNR, SSIM, and MOS were clearly improved, with a maximum PSNR of 47.690. The maximum increment of PSNR and SSIM of the secondary reconstruction images reached 21.7% and 2.8%, respectively, and the objective evaluation of the image quality was good. However, overall MOS was less than that of the single reconstruction, the perceptual quality was weakened, and the time cost was more than 130 times greater. The reconstruction effect of SRResNet was better than that of SRGAN. When SRGAN and SRResNet were used for the recognition of endocytic vesicles in Solanum nigrum root cells, the clarity of the reconstructed images was obviously improved, the boundary of the endocytic vesicles was clearer, and the number of identified endocytic vesicles increased from 6 to 9 and 10, respectively, and the mean fluorescence intensity was enhanced by 14.4% and 7.8%, respectively. Relevant research and achievements are of great significance for promoting the application of deep learning methods and image super-resolution reconstruction technology in laser confocal image studies.

The 14-3-3 protein GF14f negatively affects grain filling of inferior spikelets of rice (Oryza sativa L.).

In rice (Oryza sativa L.), later flowering inferior spikelets (IS), which are located on proximal secondary branches, fill slowly and produce smaller and lighter grains than earlier flowering superior spikelets (SS). Many genes have been reported to be involved in poor grain filling of IS, however the underlying molecular mechanisms remain unclear. The present study determined that GF14f, a member of the 14-3-3 protein family, showed temporal and spatial differences in expression patterns between SS and IS. Using GF14f-RNAi plants, we observed that a reduction in GF14f expression in the endosperm resulted in a significant increase in both grain length and weight, which in turn improved grain yield. Furthermore, pull-down assays indicated that GF14f interacts with enzymes that are involved in sucrose breakdown, starch synthesis, tricarboxylic acid (TCA) cycle and glycolysis. At the same time, an increase in the activity of sucrose synthase (SuSase), adenosine diphosphate-glucose pyrophosphorylase (AGPase), and starch synthase (StSase) was observed in the GF14f-RNAi grains. Comprehensive analysis of the proteome and metabolite profiling revealed that the abundance of proteins related to the TCA cycle, and glycolysis increased in the GF14f-RNAi grains together with several carbohydrate intermediates. These results suggested that GF14f negatively affected grain development and filling, and the observed higher abundance of the GF14f protein in IS compared with SS may be responsible for poor IS grain filling. The study provides insights into the molecular mechanisms underlying poor grain filling of IS and suggests that GF14f could serve as a potential tool for improving rice grain filling.

Microcantilever-Based Label-Free Characterization of Temperature-Dependent Biomolecular Affinity Binding.

This paper presents label-free characterization of temperature-dependent biomolecular affinity binding on solid surfaces using a microcantilever-based device. The device consists of a Parylene cantilever one side of which is coated with a gold film and functionalized with molecules as an affinity receptor to a target analyte. The cantilever is located in a poly(dimethylsiloxane) (PDMS) microfluidic chamber that is integrated with a transparent indium tin oxide (ITO) resistive temperature sensor on the underlying substrate. The ITO sensor allows for real-time measurements of the chamber temperature, as well as unobstructed optical access for reflection-based optical detection of the cantilever deflection. To test the temperature-dependent binding between the target and receptor, the temperature of the chamber is maintained at a constant setpoint, while a solution of unlabeled analyte molecules is continuously infused through the chamber. The measured cantilever deflection is used to determine the target-receptor binding characteristics. We demonstrate label-free characterization of temperature-dependent binding kinetics of the platelet-derived growth factor (PDGF) protein with an aptamer receptor. Affinity binding properties including the association and dissociation rate constants as well as equilibrium dissociation constant are obtained, and shown to exhibit significant dependencies on temperature.

Spatio-temporally resolved detection on a microfluidic chip for monitoring the dynamic processes of molecular events.

Detection is an essential aspect in analytical approaches. In liquid phase separations, many attempts have been focused on the capability to detect a partial or an entire column. However, detection in both spatial and temporal resolutions has not gained much attention yet. Here we present the concept of spatio-temporally resolved detection (STRD) and a proof-of-the-concept microchip electrophoresis (MCE)-STRD system. The MCE-STRD system was mainly composed of a microchip and an STRD unit, which were designed completely based on the requirements for spatial and temporal resolutions. In the STRD unit, a linear light beam expanded from a UV LED light source was employed to illuminate the whole separation channel of the microchip while a linear CCD sensor that has an identical effective length as the separation channel and more pixels per unit length was used to detect the absorbance signals through the separation channel. As each pixel of the CCD sensor can detect a corresponding channel space in real time, the CCD provides both spatial and temporal resolutions. A significant advantage of STRD over conventional detection schemes is its capability for monitoring the dynamic processes of molecular events occurring in the separation channel. This was demonstrated through the monitoring of the dynamic processes of protein-DNA and protein-drug interactions in chip isoelectric focusing (chip IEF). The MCE-STRD system provided not only whole pictures of the entire dynamic processes at-a-glance but also quantitative kinetic information (dissociation rate constants) of the dynamic processes. With further development, we anticipate that STRD could be a promising tool for the characterization of biomolecular interactions and the observation of migration behaviours of analytes.

A planar PDMS micropump using in-contact minimized-leakage check valves.

We present a micropump with a simple planar design featuring compliant in-contact check valves in a single layer, which allows for a simple structure and easy system integration. The micropump, based on poly(dimethylsiloxane) (PDMS), primarily consists of a pneumatically driven thin membrane, a pump chamber, and two in-plane check valves. The pair of check valves is based on an in-contact flap-stopper configuration and is able to minimize leakage flow, greatly enhancing the reliability and performance of the micropump. Systematic experimental characterization of the micropump has been performed in terms of the frequency response of the pumping flow rate with respect to factors including device geometry (e.g. chamber height) and operating parameters (e.g. pneumatic driving pressure and backpressure). The results demonstrate that this micropump is capable of reliably generating a maximum flow rate of 41 µL min-1 and operating against a high backpressure of up to 25 kPa. In addition, a lumped-parameter theoretical model for the planar micropump is also developed for accurate analysis of the device behavior. These results demonstrate the capability of this micropump for diverse applications in lab-on-a-chip systems.